The development of robust blood biomarkers of exposure and early disease manifestations can be viewed as a non-invasive alternative to improve at-risk patient identification and prognosis. Novel high-throughput techniques allow reliable measurement of the cellular activity and its regulation at different molecular levels. This defines an ideal framework to seek for biomarkers of exposures through the biochemically-imprinted effects of external exposures, and biomarkers of disease risk by identifying specific molecular alterations that arise at different stages along the course of disease progression.Previous investigations have established the existence of gene expression and epigenetic signals that are involved in the physiological responses to smoking exposure and representative of lung cancer risk. Additionally, robust evidence is accumulating to support the existence of inheritable lung cancer risk determinants. However, the way these molecular signals interact to ultimately contribute to individual lung cancer risk profile remains unknown.
In order to explore these complex regulatory cascades, Mechanomics will exploit, as a unique resource, a large dataset including full-resolution genetic profiles already available in (N>45,000) individuals, genome-wide blood-derived DNA-methylation (N>2,700) and gene expression (N= 300 are already available) data. As a new feature of the project, we will enrich this multi-OMIC dataset by including high resolution metabolomics using mass spectrometry in >1,700 participants.
The project will devise novel statistical approaches combining OMICs profiling techniques, network topologies, and longitudinal models to explore the molecular mechanisms involved in the smoking-induced cellular perturbations using the wealth of multi OMICs profiles available in healthy participants. Utilising data obtained from (nested) case control studies, we will identify which of these smoking-induced perturbations are effectively driving individual lung cancer risk. We will also use an original longitudinal model for lung cancer natural history to explicitly model genetically-driven susceptibility functions to smoking exposure and lung cancer risk. This model will be generalised to include prioritised (multi-omic) markers to predict clinical onset and the dynamics of smoking-induced lung cancer disease progression.
Altogether the proposed research will constitute the world largest multi-omic dataset for lung cancer including metabolome-wide data. This work will help identifying functionally relevant OMICs biomarkers mediating the effect of exposure to tobacco smoke towards increased risk of lung cancer, and will potentially inform both their role in lung carcinogenesis, and their ability to predict dynamics of lung cancer progression. As such Mechanomics has the potential to contribute to improved patient early diagnosis and risk stratification.